Forum Home
PC World Chat
Thread ID: 69789	2006-06-12 10:11:00	Battery Power	Twelvevolts (5457)	PC World Chat

Post ID	Timestamp	Content	User
462548	2006-06-14 03:54:00	Unless you've got a source of very cheap car batteries, don't try to run your computer off one. Car batteries are not designed for deep cycling. They are designed to briefly (5 seconds or so) supply a few hundred amps into a starter motor.The loads in the car should be supplied by the alternator. If you run a car battery down to 10.5 volts, which is where most inverters drop out (to protect the battery) it probably will not be able to start the car. And if you do manage to charge it and run it down again, you will have shortened its life severely. The gel batteries in a UPS (or the batteries from a golf cart or mobility scoooter) are designed for deep cycling, but even they live longer if they aren't taken down to an end point that low. I'm not so concerned about "autotransfomer" inverters ... I suspect they aren't around any more. Those would be iron-cored transformers: modern inverters in power supplies run at 10s of kilohertz, not 50Hz, and use miniature moulded iron dust cores to generate the peak voltage, then high voltage mosfets to create the 50Hz AC. That's why your 600W unit doesn't feel heavy, GF. The actual transformer core will be smaller than a matchbox. The change is not for safety: it's for cheapness. A wound iron cored transformer is expensive. And of course, it's expensive to ship. Howver, rather than playing around with UPSs, which are not designed to run for long, you should get a proper inverter. Small ones, which will run a laptop, are dirt cheap now. But you still need amphours. If your desktop pulls more than a hundred watts, you are looking at big amps at 12 volts. 100W would be around 10 amps from the battery. Ah ratings for lead acid batteries are "20 hour" ratings. In theory, a 17Ah battery will give less than one amp (17/20) for 20 hours. If you pulled 17 amps, you'd be looking at minutes of running time, not one hour.	Graham L (2)
462549	2006-06-14 04:26:00	Corrrect on the use of a common car battery. Installed a pair many years ago on a company 24v backup emergency lighting setup, they were "tested" daily for one hour. After a month the new batteries were stuffed, unable to handle the "deep cycles" of discharge involved, just as you note. You should have told us then .... The transformers in my UPS's are sizeable, and they are not light. Just not my idea of "600 VA" in size. I have a 150 VA unit in the car, that certainly is light and will be high frequency conversion.	godfather (25)
462550	2006-06-14 04:51:00	. . . The transformers in my UPS's are sizeable, and they are not light . Just not my idea of "600 VA" in size . . . I haven't looked inside any modern UPSs recently, the two junk ones I have lying around do appear to have double wound (iron) transformers . . . one of them had 16 12 V gel batteries in series , so you could get a good shock off the DC side . :D ) But I'd bet that modern ones would have a switchmode charger, and a high frequency inverter . The few I have seen feel to weigh about what I'd expect from the gel battery . I do have an old 600W inverter with a conventional transformer . Unfortunately, I'm not prepared to replace the two transistors which have popped, at $118 each . Its transformer is big and heavy . Your UPS ones might be lighter because of design in the US . That would indicate design (a) to run hot (=less iron, and copper) ; (b) at 60Hz (=less iron); (c) for a short period . Battery manufacturers would have told you about deep cycling . Retailers wouldn't know . I wish car manufacturers had taken to gel batteries (zero maintainence, no spills, long life) . But there's only one or two high price cars with them, so the price remains high .	Graham L (2)
462551	2006-06-17 03:27:00	I'm not so concerned about "autotransfomer" inverters ... I suspect they aren't around any more. Those would be iron-cored transformers: modern inverters in power supplies run at 10s of kilohertz, not 50Hz, and use miniature moulded iron dust cores to generate the peak voltage, then high voltage mosfets to create the 50Hz AC. That's why your 600W unit doesn't feel heavy, GF. The actual transformer core will be smaller than a matchbox. The change is not for safety: it's for cheapness. A wound iron cored transformer is expensive. And of course, it's expensive to ship. Got around to looking inside one of mine yesterday. It's less than 12 months old, and is indeed iron cored transformer based. Also there is a high voltage on the battery terminals to earth when it's running from the battery. As it's modified sime wave, it does not register well on the cheap multimeter I had to hand, and I am unsure of the impedance. But it does lend credence to the claim of aoutotransformer use, it's a single bobbin transformer with about 8 winding connections. Picture of the quite meaty transformer is attached. Its not a "matchbox size" as suggested, it occupies almost half the UPS case area and the unit is very heavy! www.imagef1.net.nz I therefore stand by my warning that connection of external batteries to a UPS not designed for them could be a very dangerous activity.	godfather (25)
462552	2006-06-17 04:05:00	And I still stand by: . . . Howver, rather than playing around with UPSs, which are not designed to run for long, you should get a proper inverter . Small ones, which will run a laptop, are dirt cheap now . . . Interesting that they still do this . I suppose it's the old rule: if a design works, don't change it . But they would have to rely on their brandname to justify the price difference . An Asian manufacturer who can start with a new design would eat them on price . Are you sure it's not just a leakage "high voltage" on the battery terminals . It seems pointless to "autowind" the heavy current 12V primary and low current AC output windings . It's no harder to keep the few turns of 10 or 12 SWG primary separate than to series connect them . 10 megohm input meters see a lot of things (particularly AC voltages) that aren't really there . What's the resistance measurement between the 12V battery terminals and the 230V ones? (With it not running at all, of course) Missed opportunity: I was talking to a friend yesterday and he asked me what a couple of things he had were . He'd been given a couple of big heavy black boxes . Two 65Ah 12V gel batteries . He thought they were "probably dead" I pulled out my meter, 11 . 85V and 12 . 6V . I told him what they were worth, and how they should be treated . I should have told him they were scrap, and that I would dispose of them for him . :D	Graham L (2)
462553	2006-06-17 07:06:00	It could be leakage, it's only present when the UPS is in battery mode. In -all power off - (including the UPS turned off) mode there appears to be no connection, it's not easy to determine (safely) when the UPS is running on battery. But the giveaway for me is that the transformer does not have the "dual bobbin" double insulated construction that should otherwise be present. Safety-wise it's OK with the regulations as there is no ability to access the device battery connections without tools and warning stickers clearly advise that the battery presents a danger of electric shock. It also has a CE compliance conformity number. As for "price difference", these cost me $87 + GST each at the time. Thats for a 600 VA UPS with smart interface. I doubt if a newer design would be cheaper than that.	godfather (25)
462554	2006-06-18 02:20:00	The other aspect which occured to me last night is that the primary winding would be push-pull, with a centre tap for supply and the switches on the ends . How would you "auto" that? Double insulated (and floating) low voltage supplies can give quite good sparks when plugged in while running . There's quite a lot of leakage -- enough to zap delicate equipment . That price is amazingly low . . . see DSE's prices for simple 230-115V transformers: 150VA $118, 1000VA $398 . That's a transfomer in a box, with no electronics, let alone a battery . ;)	Graham L (2)
462555	2006-06-18 02:39:00	It does look like a centre-tapped "push-pull". But it sure doesn't look like 600 VA in a normal transformer regarding size. Also the LV winding is straight over the HV winding, not on a separate bobbin as required for classification as "double insulated" so capacitive coupling will occur. It does not worry me, the UPS's work well (will happlily start the PC from cold if needed) but the warnings of high voltage being present on the battery terminals is there for a reason I suspect. Where electronics is required to monitor current via a shunt resistor (as in all modern electronic electricity revenue meters), they tie the LV DC supply negative to the phase potential as its simpler by far to then measure the voltage developed and translate it to current. The metering chips are ubiquitous and may well be used in these devices for simple measuring and control, which would also be a reason for caution and the warning. The chips would provide all the required parameters to control the UPS operation. If that was the case, it could also explain the lack of double insulation, it would be a waste of time and money if the negative was already at phase potential when running. The serial interface data appears to be opto-coupled in the UPS's for isolation.	godfather (25)
1 2